JP2005150506A - Semiconductor manufacturing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor manufacturing apparatus provided with a wafer holder by which the cooling speed of a heater can be improved and the uniformity of temperature distribution of the heater at the time of cooling can be secured and capable of sharply shortening the processing time of a semiconductor wafer. <P>SOLUTION: The wafer holder comprises the heater 1 on which a semiconductor wafer is set and heated and a cooling block 2 for cooling the heater 1. The cooling block 2 is movably arranged so as to be abutted on and separated from a rear face 1b which is the opposite side of the wafer mounting surface 1a of the heater 1, and the warp of the abutting surface 2a which is abutted on the heater 1 is ≤1 mm. A passage for a cooling solution can be formed in the cooling block 2 and it is preferable that the cross-sectional area of the passage is ≥1 mm<SP>2</SP>in 80% and more of the whole length of the passage and the area of a part on which the passage is formed is ≥3% of the whole area of the abutting surface 2a. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a semiconductor manufacturing apparatus such as plasma CVD, low pressure CVD, metal CVD, insulating film CVD, ion implantation, etching, low-K film formation, DEGAS apparatus and coater developer, particularly a semiconductor wafer for carrying out a predetermined process. The present invention relates to a semiconductor manufacturing apparatus provided with a wafer holder for mounting and heating.

  Conventionally, there are many processes such as various film formation, etching, and photolithography as a process applied to a semiconductor wafer. In these processes, a semiconductor wafer is placed on a wafer holder provided with a heater in a semiconductor manufacturing apparatus, and film formation and other processes are performed while heating. Therefore, various wafer holders suitable for each process have been studied.

  For example, in a photolithography process for forming a resist film pattern on a wafer, the wafer is washed, heated and dried, cooled, and then coated with a resist solution on the wafer surface and dried, followed by exposure, development, etc. Is processed. By the way, in this photolithography process, since the quality of the coating film is greatly influenced by the temperature at which the resist is dried, it is important to keep the temperature at the time of processing uniform and constant. Further, the processing apparatus using a metal heater has a problem that a large number of particles adhere to the semiconductor wafer.

  Further, the processing of these semiconductor wafers needs to be completed in as short a time as possible in order to improve the throughput as much as possible. In response to such a requirement, the present inventors have provided a semiconductor in which a block portion that can be contacted and separated is disposed on the back surface of the heater opposite to the wafer mounting surface in order to shorten the wafer processing time. A heater module for a manufacturing apparatus has been proposed (see Japanese Patent Application No. 2002-163747).

  In order to shorten the processing time of the semiconductor wafer, as described above, the wafer holding body that performs various processes while placing and heating the semiconductor wafer is brought into contact with and separated from the back surface of the heater opposite to the wafer mounting surface. It is effective to use a wafer holder to which a block capable of being attached is attached. By using such a wafer holder, the cooling rate of the heater can be remarkably improved, and the processing time of the semiconductor wafer can be shortened.

  However, it was newly found that when the cooling block is brought into contact with the heater of the wafer holder, the temperature distribution of the heater becomes non-uniform from the start of cooling to the completion of cooling. As described above, the use of the wafer holder in which the temperature distribution of the heater during cooling is not uniform is limited to the use in which the uneven temperature distribution does not become an obstacle. Further, when the temperature distribution of the heater at the completion of cooling becomes non-uniform, the time until the uniformity of the temperature distribution of the heater is restored after the cooling is completed increases, so that the shortening of the processing time of the semiconductor wafer is offset.

  In view of such circumstances, the present invention improves the cooling rate of a wafer holder that performs various processes while placing and heating a semiconductor wafer, and at the same time, the heater from the start of cooling to the completion of cooling. It is an object of the present invention to provide a semiconductor manufacturing apparatus that can ensure the uniformity of the temperature distribution of the semiconductor wafer and can significantly reduce the processing time of the semiconductor wafer by providing such a wafer holder.

  In order to achieve the above object, a semiconductor manufacturing apparatus provided by the present invention includes a wafer holder including a heater for placing and heating a semiconductor wafer and a cooling block for cooling the heater, and includes a cooling block. Is arranged to be movable so as to be able to contact and separate from the back surface of the heater opposite to the wafer mounting surface, and the warpage of the contact surface contacting the heater of the cooling block is 1 mm or less. The warpage of the contact surface that contacts the heater of the cooling block is preferably 0.2 mm or less, and more preferably 0.05 mm or less.

  In the semiconductor manufacturing apparatus of the present invention, when the cooling block moves and contacts the heater, the angle formed between the contact surface of the cooling block and the back surface of the heater is preferably 10 ° or less. In the semiconductor manufacturing apparatus of the present invention, it is preferable that a chamfering process of 10 μm or more is performed on a corner portion of the contact surface that contacts the heater of the cooling block.

  In the semiconductor manufacturing apparatus of the present invention, when the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is preferably 10 ° or less. In the semiconductor manufacturing apparatus of the present invention, a distance from the center of the wafer mounting surface to the outer edge has a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface. In an intersecting line where a cylindrical surface having a shorter radius and an arbitrary plane perpendicular to the wafer mounting surface and passing through the center of the wafer mounting surface intersect, the wafer mounting surface on the intersection line The length to the inner surface of the container is preferably in the range of 0.9 to 1.1 times the average value of the length.

The semiconductor manufacturing apparatus of the present invention may include a cooling liquid flow path in the cooling block. In the semiconductor manufacturing apparatus of the present invention provided with this flow path, it is preferable that 80% or more of the total length of the flow path has a flow path cross-sectional area of 1 mm ² or more. The area of the portion where the flow path is formed is preferably 3% or more of the total area of the contact surface when viewed from a direction perpendicular to the contact surface with which the heater of the cooling block contacts. .

  In the semiconductor manufacturing apparatus of the present invention provided with the above-described flow path, the flow path is located on the inner side of the outer edge of the abutment surface, when viewed from a direction perpendicular to the abutment surface with which the heater of the cooling block abuts. It is preferable to form in the range. Moreover, it is preferable that the surface roughness of the surface which the liquid of the said channel contacts is 0.02-100 micrometers in Ra. Furthermore, the flow rate of the liquid flowing through the flow path is preferably 500 cc / min or more.

  In the above-described semiconductor manufacturing apparatus of the present invention, the thermal conductivity of the material constituting the cooling block is preferably 30 W / mK or more, and more preferably 100 W / mK or more.

  In the semiconductor manufacturing apparatus of the present invention described above, the main component of the material constituting the heater is preferably aluminum nitride, aluminum oxide, silicon carbide, or silicon nitride, and the material constituting the heater is mainly used. More preferably, it is made of aluminum nitride.

  According to the present invention, with respect to a wafer holder that performs various processes while placing and heating a semiconductor wafer, the cooling rate of the heater is improved and at the same time, the temperature distribution of the heater is uniform from the start of cooling to the completion of cooling. Sex can be secured. Therefore, it is possible to provide a semiconductor manufacturing apparatus that includes this wafer holder and can significantly reduce the processing time of the semiconductor wafer.

  In the semiconductor manufacturing apparatus of the present invention, as shown in FIG. 1 and FIG. 2, as a wafer holder for performing various processes while placing and heating a semiconductor wafer, the semiconductor wafer is placed and heated. A wafer holder comprising a heater 1 and a cooling block 2 for cooling the heater 1 is disposed in a container 3 of the apparatus. A thermocouple 4 and energizing wires 5 and 5 are attached to the heater 1 through the through hole of the cooling block 2 in the wafer holder, and the heater 1 is heated to a predetermined temperature.

  The cooling block 2 is movably arranged by a plurality of drive shafts 6 and 6. Then, by driving the drive shafts 6 and 6, the cooling block 2 is brought into contact with the heater back surface 1b opposite to the wafer mounting surface 1a of the heater 1 (state shown in FIG. 2), or the wafer mounting surface 1a. Can be separated from the back surface 1b of the substrate (the state of FIG. 1). For example, when the heater is heated and kept at a high temperature, the cooling block 2 is separated from the heater 1 as shown in FIG. 1, and when the heater is cooled, the cooling block 2 is brought into contact with the heater 1 as shown in FIG. By doing so, the cooling rate of the heater 1 can be significantly improved.

  As described above, in the wafer holder according to the present invention including the heater and the cooling block, the warping of the contact surface 2a (the surface contacting the heater back surface 1b) of the cooling block 2 is 1 mm or less. Setting the warpage of the contact surface 2a of the cooling block 2 to 1 mm or less is effective in making the temperature distribution of the heater 1 uniform during cooling (from the start of cooling to the completion of cooling). When the cooling rate is further increased, the warping of the contact surface of the cooling block is preferably 0.2 mm or less, and more preferably 0.05 mm or less. In general, the warpage of a metal plate having a diameter of about 300 mm is usually more than 1 mm with aluminum or its alloy having excellent thermal conductivity. In a conventional cooling block using such a metal plate, a heater for cooling is used. There was a large variation in the temperature distribution.

  As described above, when the warpage of the contact surface 2a of the cooling block 2 is 1 mm or less, the cooling block 2 and the heater 1 are in contact with each other and heat is uniformly transmitted from the heater 1 to the cooling block 2. It can cool uniformly. That is, when the warpage of the contact surface 2a of the cooling block 2 exceeds 1 mm, a portion where the cooling block 2 contacts the heater 1 and a portion where the cooling block 2 does not contact are generated. At this time, since the cooling rate inevitably decreases in the portion of the heater 1 that does not contact the cooling block 2, the temperature distribution of the heater 1 varies greatly. Further, when the warpage of the contact surface of the cooling block 2 is 1 mm or less, the contact area between the cooling block 2 and the heater 1 is substantially increased, so that the cooling rate of the heater 1 can also be improved. .

  Further, in the wafer holder according to the present invention, when the cooling block moves and contacts the heater, that is, at the moment when the heater and the cooling block come into contact, the angle formed between the heater back surface and the contact surface of the cooling block is 10. It is preferable to adjust to less than or equal to. The heater temperature distribution is most likely to be non-uniform at the moment when the cooling block comes into contact. However, if the angle between the heater back surface and the contact surface of the cooling block is 10 ° or less, the contact timing will be within the contact surface. It is possible to effectively suppress the nonuniformity of the temperature distribution caused by the deviation.

  Furthermore, it is preferable to chamfer 10 μm or more at the corner of the contact surface with which the heater of the cooling block contacts. Burrs are likely to occur at the corners of the cooling block. If these burrs are present at the corners of the contact surface with the heater, complete contact between the heater back surface and the contact surface of the cooling block is hindered. Therefore, by removing the burrs that exist at the corners of the contact surface of the cooling block by chamfering, extremely high uniformity is obtained in the contact between the back surface of the heater and the contact surface of the cooling block. The uniformity of temperature distribution and the cooling rate can be further improved.

  In addition, when the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is preferably 10 ° or less. Depending on the use and use state of the heater, non-uniform temperature distribution of the heater caused by a slight difference in distance between the heater and the cooling block becomes a problem. That is, in the portion where the distance between the heater and the cooling block is short, more heat escapes from the heater to the cooling block through the gas existing between the heater and the cooling block than in the portion where the distance is long. The temperature of the heater of the part falls. By setting the angle formed by the contact surface of the cooling block and the back surface of the heater to 10 ° or less, this non-uniform temperature distribution can be effectively suppressed.

  Further, the same relationship between the contact surface of the cooling block and the back surface of the heater can be established between the wafer mounting surface of the heater and the inner surface of the container of the semiconductor manufacturing apparatus. That is, ideally, the temperature distribution of the heater can be made more uniform by setting the angle formed by the wafer mounting surface of the heater and the inner surface of the container of the semiconductor manufacturing apparatus to 10 ° or less. However, the inner surface of the container of the semiconductor manufacturing apparatus that faces the wafer mounting surface of the heater is not necessarily flat because of limitations due to the design of the entire apparatus. Therefore, when the inner surface of the container is not flat, it has a center line that is perpendicular to the wafer mounting surface of the heater and passes through the center of the wafer mounting surface, and is shorter than the distance from the center of the wafer mounting surface to the outer edge. In an intersecting line where a cylindrical surface having a radius intersects an arbitrary plane perpendicular to the wafer mounting surface and passing through the center of the wafer mounting surface, the inner surface of the container of the semiconductor manufacturing apparatus extends from the wafer mounting surface on the intersecting line. The length up to 0.9 may be 1.1 to 1.1 times the average value of the length.

  By setting in this way, the distance between the wafer mounting surface of the heater and the inner surface of the container of the semiconductor manufacturing apparatus has a concentric distribution, and the gas existing between the wafer mounting surface and the inner surface of the container is reduced. Since the way of heat escape through is also concentric, the heater temperature distribution is also concentric. This concentric temperature distribution can be easily eliminated by designing the heater circuit pattern of the heater. For example, in a concentric portion where the temperature is low, by reducing the wiring width of the heater circuit pattern of the heater or reducing the wiring interval, the heat generation density of that portion can be increased and the temperature can be raised. On the other hand, in the concentrically high temperature portion, the heat generation density of the portion can be reduced by increasing the wiring width of the heater heating circuit pattern or by increasing the wiring interval, thereby lowering the temperature. Therefore, even when the inner surface of the container of the semiconductor manufacturing apparatus is not flat, the temperature distribution of the heater can be easily made uniform.

  In the wafer holder of the present invention having the above-described configurations, the higher the temperature uniformity, the higher the cost for realizing the configuration. Therefore, these configurations may be appropriately selected and adopted according to the use and purpose of the wafer holder.

  With respect to the cooling block in the wafer holder of the present invention, a cooling liquid can be allowed to flow in the interior for the purpose of improving the cooling capacity and throughput. By flowing the liquid into the cooling block, the cooling capacity can be further improved and the cooling rate of the heater can be further increased. In addition, when the processing of the semiconductor wafer is continuously performed, even if the cooling of the heater is continuously performed in the cooling block, heat is gradually accumulated in the cooling block with the implementation, and the cooling capacity is reduced. Even in such a case, it is possible to prevent the cooling capacity of the cooling block from gradually decreasing by flowing the liquid through the cooling block.

  Although there is no restriction | limiting in particular as a cooling liquid sent through the inside of a cooling block, Organic solvents, such as a fluorinate, can be used other than water. In order to improve the cooling capacity, the liquid can be cooled with a chiller or the like. In addition, when flowing the cooling liquid in the cooling block, the flow path through which the liquid flows may be one system, or a plurality of systems may be provided. When a plurality of channels are provided to flow liquid, it is possible to provide a higher cooling capacity than in the case of only one channel. However, since the apparatus becomes complicated by providing a plurality of channels, it is desirable to use them according to the purpose and purpose.

  When the liquid is allowed to flow inside the cooling block, it is natural that the cooling block is brought into contact with the heater when the heater is cooled, but the cooling block can always be brought into contact with the heater even during cooling. With respect to the timing of flowing the liquid into the cooling block, the liquid may or may not flow regardless of whether or not it is in contact with the heater when the heater is heated and kept at a high temperature. When the liquid is not flowed, the temperature of the peripheral parts of the heater rises, and depending on the parts, heat resistance is required. However, the heating rate of the heater is increased, and the power consumption can be reduced. However, since the temperature of the cooling block rises as the heater temperature rises, the cooling rate slightly decreases even when the liquid is flowed after the heater is cooled. Therefore, in consideration of such points, it is necessary to appropriately select whether or not the liquid is allowed to flow even when the heater is heated and when it is kept.

  In addition, when flowing liquid into the cooling block, not only the contact between the heater and the cooling block, but also the uniformity of the temperature distribution of the cooling block, the temperature distribution of the heater from the start of cooling to the completion of cooling is further improved. It can be made uniform. That is, when the temperature of the cooling block is non-uniform, the non-uniform temperature distribution is naturally reflected in the heater cooled by the cooling block, leading to non-uniformity of the heater temperature distribution. On the contrary, if the temperature distribution of the cooling block can be made uniform, the uniformity of the temperature distribution of the heater during cooling can be further improved.

As a means for improving the uniformity of the temperature distribution of the cooling block through which the liquid flows, first, the cross-sectional area of the flow path through which the liquid flows is preferably 1 mm ² or more over 80% or more of the total length of the flow path. Since the temperature of the liquid flowing in the cooling block flows while taking heat from the cooling block, the temperature of the liquid is higher at the outlet of the channel than at the inlet of the channel. The temperature difference of the liquid between the inlet and outlet of this flow path is one of the causes that cause the uneven temperature distribution of the cooling block. Therefore, by setting 80% or more of the total length of the flow path of the liquid to a flow path having a cross-sectional area of 1 mm ² or more, a sufficiently large flow rate can be secured for the liquid flowing in the cooling block. As a result of reducing the temperature difference of the liquid between the outlets, the uniformity of the temperature distribution of the cooling block can be improved.

  As another means for improving the uniformity of the temperature distribution of the cooling block through which the liquid flows, the area of the portion where the flow path is formed when viewed from the direction perpendicular to the contact surface that contacts the heater of the cooling block However, it is effective to form the flow path so as to be 3% or more of the area of the contact surface of the cooling block. Even if the area of the flow path on the contact surface of the cooling block is small, a certain degree of temperature distribution uniformity can be ensured by the thermal conductivity of the material constituting the cooling block itself, but it is relatively far from the position close to the flow path. A slight temperature difference will occur depending on the position. If the area of the flow path occupying the contact surface of the cooling block is 3% or more, the temperature difference due to the position from the flow path can be eliminated, and higher uniformity of the temperature distribution of the cooling block can be realized.

  Furthermore, as another means for improving the uniformity of the temperature distribution of the cooling block through which the liquid flows, when viewed from the direction perpendicular to the contact surface that contacts the heater of the cooling block, the inner side of the outer surface of the contact surface is less than 50 mm. It is preferable to form the flow path in the range. In this way, by forming a flow path from the center of the cooling block to the outer edge and flowing the liquid, it is possible to cool the entire cooling block with the liquid evenly, so the temperature distribution of the cooling block is uniform. It is possible to further improve the property.

  The flow path for flowing the liquid to the cooling block can be formed by, for example, the following method. First, a groove to be a flow path is formed on the surface of one block by counterboring or the like, and a groove for inserting a sealing material such as an O-ring is formed in the vicinity of the outer edge side of the flow path. After inserting an O-ring or the like into this groove, another block is superposed on the surface of this block and tightened with a screw or the like to produce a cooling block. This method is desirable in that the shape of the flow path can be freely produced by processing. Moreover, a cooling block can also be produced by joining two blocks, each having a channel serving as a flow path, on the surface by welding or brazing.

  Regarding the flow path formed in this way, from the viewpoint of the cooling capacity of the cooling block, the surface roughness of the surface in contact with the liquid in the flow path is Ra (center line average roughness) in the range of 0.02 to 100 μm. It is good to do. When the surface roughness Ra is less than 0.02 μm, the contact area between the liquid and the flow path is reduced, and the liquid cannot efficiently remove heat from the cooling block, leading to a decrease in the cooling capacity of the cooling block. . On the other hand, when the surface roughness Ra is larger than 100 μm, the resistance when the liquid passes through the flow path is increased, and the flow rate and flow rate of the liquid are decreased. Since the temperature difference of the liquid increases between the inlet and the outlet of the flow path, the uniformity of the temperature distribution is deteriorated. In particular, when the surface roughness Ra is about 10 μm, the contact area between the liquid and the flow path and the balance of the flow rate and flow rate of the liquid are good, and high performance in both cooling capacity and uniformity of temperature distribution can be obtained.

  The flow rate of the liquid flowing through the flow path in the cooling block is desirably 500 cc / min or more. Even if the flow rate is smaller than this, the heater can be cooled effectively, but by setting the flow rate to 500 cc / min or more, not only can a higher cooling capacity be realized, but also between the inlet and outlet of the flow path. The temperature distribution non-uniformity caused by the temperature difference between the liquids can be more effectively suppressed.

  Regarding the material of the cooling block, unlike the heater, the cooling block is not in direct contact with the wafer, so that the adhesion of particles is not a problem, and can therefore be made of metal. However, from the viewpoint of improving the uniformity of the temperature distribution of the heater, the thermal conductivity of the cooling block is also important. In this respect, the uniformity of the temperature distribution of the heater can be kept high by using a material having a thermal conductivity of 30 W / mK or more. In applications that require particularly high uniformity in the temperature distribution of the heater, extremely high uniformity can be realized with respect to the temperature distribution of the heater by using a material having a thermal conductivity of 100 W / mK or more.

  On the other hand, various heater materials are conceivable. However, when a metal is used, there is a problem that a large number of particles adhere to the wafer. Among ceramics, aluminum nitride and silicon carbide having high thermal conductivity are preferable in order to obtain uniformity of temperature distribution. If importance is placed on reliability, silicon nitride is preferable because it is strong and resistant to thermal shock. If importance is attached to the cost, aluminum oxide is preferable. Among these, aluminum nitride is particularly preferable in consideration of the balance between performance and cost.

Next, a method for manufacturing the heater will be described by taking a particularly preferable aluminum nitride heater as an example. The specific surface area of the aluminum nitride powder used as a raw material is preferably 2.0 to 5.0 m ² / g. When the specific surface area is smaller than 2.0 m ² / g, the sinterability of aluminum nitride is lowered, which is not preferable. On the other hand, when the specific surface area is larger than 5.0 m ² / g, the aggregation of the powder becomes very strong, which is not preferable. The amount of oxygen contained in the aluminum nitride raw material powder is preferably 2% by weight or less. If the amount of oxygen exceeds 2% by weight, the thermal conductivity of the sintered body decreases, which is not preferable. The total amount of metal impurities excluding aluminum contained in the raw material powder is preferably 2000 ppm or less. When the amount of metal impurities exceeds 2000 ppm, the thermal conductivity is lowered, which is not preferable. Furthermore, group IV elements such as Si and iron group elements such as Fe are not particularly preferred as metal impurities because they have a high effect of reducing the thermal conductivity, and the content is preferably 500 ppm or less.

  Since aluminum nitride is a difficult-to-sinter material, it is preferable to add a commonly used sintering aid. These sintering aids react with the aluminum oxide or aluminum oxynitride present on the surface of the aluminum nitride particles to promote densification of the aluminum nitride and contribute to lowering the thermal conductivity of the aluminum nitride. Since it traps certain oxygen, it works to improve thermal conductivity. The addition amount of the sintering aid is preferably in the range of 0.01 to 5.0% by weight in terms of oxide. If the addition amount of the sintering aid is less than 0.01% by weight, not only is it difficult to obtain a sufficiently dense sintered body, but also the thermal conductivity is lowered, which is not preferable. Further, if the amount of the sintering aid added exceeds 5.0% by weight, when the heater is used in a corrosive atmosphere, the sintering aid present at the grain boundary is etched in the corrosive atmosphere. This is not preferable because it causes particles.

  As the sintering aid, a rare earth element compound is preferable, and an yttrium compound having a high trapping capability for removing oxygen from aluminum nitride is particularly preferable. As the form of the rare earth element compound, oxide, nitride, fluoride, stearic acid compound, etc. can be used. Of these compounds, oxides have the advantage of being particularly inexpensive and readily available. The stearic acid compound is particularly suitable because it has a high affinity with an organic solvent, and particularly when the raw material powder and the sintering aid are mixed in an organic solvent, the mixing property becomes high.

  To these raw material powders, a predetermined amount of a solvent, a binder and, if necessary, a dispersant and a glaze are added. These are mixed to form a slurry. As the mixing means, ball mill mixing or ultrasonic mixing is possible. The obtained slurry is formed into a sheet by a doctor blade method. Although there is no restriction | limiting in particular when shape | molding a sheet | seat, About the thickness of a sheet | seat, it is preferable that the thickness after drying shall be 3.0 mm or less. This is because if the thickness of the sheet exceeds 3.0 mm, the amount of drying shrinkage of the slurry increases, and the possibility of cracks in the sheet increases.

On the other hand, the conductor paste used for forming the heating element circuit pattern is preferably a paste of a refractory metal such as tungsten, molybdenum, or tantalum because of the thermal expansion coefficient matching with ceramics such as aluminum nitride constituting the heater. These powders are sufficiently mixed, and a binder and a solvent are added to produce a conductor paste. Moreover, in order to ensure adhesion strength with the aluminum nitride sintered body, it is possible to add an oxide to the paste. The oxide to be added is not particularly limited as long as it wets both the aluminum nitride and the refractory metal, and specifically, oxides of Group IIA and IIIA elements, Al ₂ O ₃ and SiO ₂ are preferable. In particular, yttrium oxide is preferable because it has very good wettability with respect to aluminum nitride. The amount of these oxides added is preferably 30% by weight or less.

  Next, in the case of the cofiring method (cometallizing method) in which the heating element circuit pattern is sintered simultaneously with the sheet, the heating element circuit pattern is formed on the sheet by screen printing using this conductive paste. The dry film thickness of the heating element circuit pattern (metal layer) at this time is preferably 5 to 100 μm. A film thickness of less than 5 μm is not preferable because the resistance value becomes too high and the adhesion strength becomes low. On the other hand, if the film thickness exceeds 100 μm, the adhesion strength is lowered, which is not preferable. At the same time as the formation of the heating element circuit pattern, an RF electrode, an electrostatic chuck electrode, and the like can be formed by screen printing as necessary.

  When the heater is used in a corrosive atmosphere, it must be embedded in aluminum nitride so that the heating element is not corroded. For this reason, another sheet on which the circuit is not formed is laminated on the sheet on which the circuit is formed as described above, and the heating element circuit pattern is embedded. For example, the sheets are superposed so as to sandwich the heating element circuit pattern, a solvent is applied between the sheets as necessary, and pressure is applied while heating to integrate the sheets. Since the heating temperature affects the flexibility of the finished sheet, it may be unnecessary, but is generally preferably 150 ° C. or lower. This is because if the temperature higher than this is applied, the sheet is greatly deformed. As the pressure to be applied, if the pressure is less than 1 MPa, the sheets do not sufficiently adhere to each other, so that so-called delamination is likely to occur in steps such as degreasing and sintering, and if the pressure exceeds 100 MPa, the deformation amount of the sheet becomes too large. A range of 100 MPa is preferred.

  The obtained laminate or the sheet on which the heating element circuit pattern is formed is degreased at about 500 to 1000 ° C. in a non-oxidizing atmosphere and then sintered at a temperature of about 1700 to 2000 ° C. Examples of the non-oxidizing atmosphere gas include nitrogen and argon. Nitrogen is particularly preferable because it is relatively inexpensive. Moreover, as a moisture content contained in the nitrogen used, it is preferable that a dew point is -30 degrees C or less. This is because when it contains more moisture, oxynitride is formed by the reaction between aluminum nitride and moisture in the atmosphere during sintering, which may cause a decrease in thermal conductivity. Also, the amount of oxygen contained in nitrogen is preferably 0.001% or less for the same reason as described above. As a jig used for sintering, a boron nitride (BN) compact is suitable. This BN compact not only has a heat resistant temperature against sintering, but also has a solid lubricity on the surface, so that when the compact shrinks due to sintering, the friction between the jig, the compact and the laminate is reduced. This is particularly preferable because a sintered body with less deformation can be obtained.

In the case where the heating element is not embedded in aluminum nitride with respect to the obtained sintered body, in order to ensure insulation from the cooling block to be used, other than the part connecting the power supply terminal is an insulator. Can be coated. The coating material is not particularly limited as long as it is an insulator, has low reactivity with the heating element, and has a thermal expansion coefficient difference of 5.0 × 10 ⁻⁶ / K or less with aluminum nitride. For example, crystallized glass or aluminum nitride can be used. These materials are made into a paste, for example, an insulating film having a predetermined film thickness is formed by screen printing, degreased in a non-oxidizing atmosphere at 500 to 1000 ° C., and then fired at a predetermined temperature to form a coating. Is possible.

  Since the sintered body is used as a wafer holder of a semiconductor manufacturing apparatus, it is processed as necessary. As processing accuracy, for example, the flatness of the wafer mounting surface is preferably 0.5 mm or less, and more preferably 0.1 mm or less. If the flatness of the wafer mounting surface exceeds 0.5 mm, a gap is likely to be formed between the wafer and the mounting surface of the heater. Therefore, temperature unevenness is likely to occur when the heat generated by the heater is transferred to the wafer. Therefore, it is not preferable. Further, the surface roughness of the wafer mounting surface is preferably 5 μm or less in terms of Ra, and more preferably 1 μm or less. When the surface roughness Ra exceeds 5 μm, the ceramics may be more likely to fall out of the heater due to friction between the heater and the wafer. In this case, the dropped particles become particles, which adversely affect processing such as film formation and etching on the wafer. Needless to say, it is more preferable that the surface roughness Ra is 1.0 μm or less.

  As a method for producing the heater, there is a post-fire method (post-metallization method) in which the heat-generating body is baked on pre-sintered aluminum nitride, in addition to the co-fire method (co-metallization method) in which the heat-generating body and the ceramic are simultaneously fired. The post-fire method (post metallization method) will be described below.

For example, a slurry containing aluminum nitride and a sintering aid is produced by the same method as that produced above. Granules are produced from this slurry by a spray dryer or the like. This granule is inserted into a mold having a predetermined shape, and press molding is performed. The pressing pressure at this time is preferably 9.8 MPa or more. When the pressure is less than this, the strength of the molded body is often not sufficient, and is easily damaged by handling. Although the density of a molded object changes with binder content and auxiliary agent addition amount, it is preferable that it is 1.5 g / cm < ³ > or more. If the density of the molded body is less than this, the interparticle distance becomes relatively large, and sintering is difficult to proceed, which is not preferable. Moreover, it is preferable that a molded object density is 2.5 g / cm < ³ > or less, and when it exceeds this, it will become difficult to fully remove the binder component in a molded object at the time of degreasing. For this reason, excess carbon and a carbon compound remain in the molded body, which inhibits the sintering of aluminum nitride, so that a sintered body having a sufficient density cannot be obtained.

  This molded body is degreased in a non-oxidizing atmosphere. When the molded body is degreased in an oxidizing atmosphere such as air, the surface of the aluminum nitride powder is oxidized, which causes a decrease in the thermal conductivity of the sintered body. The degreasing treatment temperature is preferably 500 to 1000 ° C. When degreasing is performed at a temperature lower than this range, the binder component cannot be sufficiently removed, and an excessive amount of carbon is present in the molded body, thereby inhibiting sintering. That is, if the amount of carbon present in the molded body after completion of degreasing exceeds 1.0% by weight, the carbon and the carbon compound in the molded body hinder the sintering of aluminum nitride, which is not preferable. Sintering of the degreased molded body is performed at a temperature of about 1700 to 2000 ° C. in a non-oxidizing atmosphere, as in the case of the above-mentioned cofiring method.

  The obtained sintered body is processed as necessary. That is, in order to form a heating element circuit pattern by screen printing in the subsequent process, it is preferable that the surface roughness Ra is 5.0 μm or less. When the surface roughness is higher than this, pattern blurring and printing defects such as pinholes are likely to occur when a circuit is formed by screen printing. The surface roughness Ra is more preferably 1.0 μm or less. Moreover, when processing a sintered compact, it is preferable to process not only a printing surface but both main surfaces. When printing is performed on one main surface and only the printed surface is polished, the unpolished surface supports the semiconductor wafer. For this reason, if there are protrusions or foreign matters on the unpolished surface, the fixing of the wafer becomes unstable. Therefore, it is preferable to process both surfaces within the above-mentioned surface roughness range.

  Further, the parallelism of both processed surfaces of the sintered body is preferably 0.5 mm or less, and more preferably 0.1 mm or less. When the parallelism exceeds 0.5 mm, the variation in film thickness may be increased when the heating element circuit pattern is screen-printed, which is not preferable. Furthermore, the flatness of the printed surface is preferably 0.5 mm or less, and more preferably 0.1 mm or less. This is because when the flatness of the printed surface exceeds 0.5 mm, the film thickness tends to vary during screen printing of the heating element circuit pattern.

Thereafter, a heating element circuit pattern is formed on the sintered body that has been processed by screen printing. The conductor paste used is the same as in the case of the cofire method. That is, as the metal powder, tungsten, molybdenum, tantalum or the like is preferable. In order to ensure adhesion strength with the aluminum nitride sintered body, oxides of Group IIA and Group IIIA elements, oxides such as Al ₂ O ₃ and SiO ₂ are used. Can be added. The amount of these oxides added is preferably 0.1 to 30% by weight. This is because if the oxide content is less than 0.1% by weight, the effect of improving the adhesion strength with the ceramic is small, and if it exceeds 30% by weight, the resistance value of the metal layer increases. The dry film thickness of the heating element circuit (metal layer) is preferably 5 to 100 μm. A film thickness of less than 5 μm is not preferable because the resistance value becomes too high and the adhesion strength becomes low. On the other hand, if the film thickness exceeds 100 μm, it is also not preferable because it causes a decrease in adhesion strength. In addition to the formation of the heating element circuit pattern, the RF electrode and the electrostatic chuck electrode can be formed by screen printing as necessary.

  The formed heating element circuit pattern is degreased in a non-oxidizing atmosphere. The treatment temperature is preferably 500 ° C. or higher. If the temperature is lower than this, carbon remains in the formed metal layer, and a refractory metal and carbide may be formed when fired. The subsequent firing is performed at 1500 ° C. or higher in a non-oxidizing atmosphere. If the firing temperature is lower than this, the grain growth of the refractory metal powder does not proceed, so the resistance value becomes very high, which is not preferable. The firing temperature should not exceed the firing temperature of the ceramic. When the metal layer is fired at a temperature exceeding the firing temperature of the ceramic, the sintering aid contained in the ceramic begins to evaporate, and further, the grain growth of the metal powder in the metal layer is promoted, and the ceramic This is because the adhesion strength to the lowers.

  In order to ensure insulation between the obtained heating element circuit patterns, it is also possible to form an insulating coating on the heating element circuit patterns. As the material of the insulating material used, if the ceramic composition of the sintered body and the composition of the insulating coat differ significantly, the coefficient of thermal expansion will of course differ, and warpage will occur after firing. It is preferable to use a material. For example, when the sintered body is aluminum nitride, a predetermined amount of Group IIA and Group IIIA oxides and carbonates are added to aluminum nitride as a sintering aid and mixed, and a binder and solvent are further added to form a paste. Apply and sinter by printing. The amount of sintering aid at this time is preferably 0.01% by weight or more in order to ensure densification and ensure insulation between the heating element circuit patterns. On the other hand, if the amount of sintering aid exceeds 20% by weight, excess sintering aid penetrates into the metal layer and changes the resistance value of the heating element, which is not preferable. The film thickness of the insulating coat is preferably 5 μm or more in order to obtain the desired insulating properties.

  It is also possible to join another sintered body to the aluminum nitride sintered body in which the heating element circuit pattern is formed and the insulating coating is formed as necessary. Thus, by laminating and bonding the sintered bodies, for example, even when a heater is used in a corrosive atmosphere or an oxidizing atmosphere, corrosion and oxidation of the heating element can be prevented. As a joining method, IIA and IIIA group oxides and carbonates are added to aluminum nitride powder, and a binder and solvent are further added to form a paste, which is then applied to the joining surface of the sintered body by a method such as screen printing. To do. Although there is no restriction | limiting in particular in the thickness of the joining layer at this time, It is preferable that it is 5 micrometers or more. This is because, if the film thickness is less than 5 μm, the bonding layer is thin, so that bonding defects are likely to occur.

  The applied paste is degreased at a temperature of 500 ° C. or higher in a non-oxidizing atmosphere. Thereafter, the two sintered bodies to be joined are overlapped with each other, applied with a predetermined load, and heated and joined in a non-oxidizing atmosphere. The load amount at this time is preferably 4.9 kPa or more, and if it is less than this, sufficient bonding strength cannot be obtained, and bonding defects such as pinholes and bonding unevenness are likely to occur, which is not preferable. In addition, the bonding temperature may be a temperature at which the sintered body and the bonding layer can sufficiently adhere to each other, and is preferably 1500 ° C. or higher. Below this temperature, sufficient bonding strength cannot be obtained and bonding defects are likely to occur, which is not preferable. Further, the bonding atmosphere is preferably performed in a non-oxidizing atmosphere such as nitrogen or argon in order to prevent the ceramics from being oxidized.

  It is also possible to form an insulating layer using an insulator such as crystallized glass after forming a heating element circuit pattern on the sintered body. In this case, a binder or an organic solvent is added to a predetermined glass powder to form a paste. This paste is applied by screen printing so as to cover the heating element circuit pattern, degreased, and then fired at a predetermined temperature.

An aluminum nitride (AlN) heater was produced by a post metallization method. That is, 0.6 parts by weight of yttrium stearate as a sintering aid was added to 100 parts by weight of AlN powder having a specific surface area of 3.4 m ² / g and an average particle size of 0.6 μm, and an organic solvent and A binder was added and granules were prepared by spray drying. This granule was press-molded, degreased at 70 ° C. in a nitrogen atmosphere, and then sintered at 1850 ° C. in a nitrogen atmosphere to produce an aluminum nitride sintered body. The obtained aluminum nitride sintered body was processed into a diameter of 330 mm and a thickness of 12 mm.

Next, 1.0 part by weight of yttrium oxide powder was added to 100 parts by weight of tungsten powder as a conductor paste for heating element formation, and a binder and an organic solvent were added to form a paste. This conductor paste was applied on the aluminum nitride sintered body by a screen printing method, degreased at 900 ° C. in a nitrogen atmosphere, and fired at 1800 ° C. in a nitrogen atmosphere. A ZnO—B ₂ O ₃ —Al ₂ O ₃ based glass paste was applied to the thickness of 100 μm except for the power feeding portion, and baked at 700 ° C. in a nitrogen atmosphere. In addition, an aluminum nitride heater was completed by attaching a W terminal with Au solder to the power supply portion, and further screwing a nickel electrode to the W terminal.

  On the other hand, two aluminum plates having a diameter of 330 mm and a thickness of 12 mm and 7 mm were prepared as cooling blocks. Of these, as shown in FIG. 3, a 12 mm thick aluminum plate was provided with a flow path 7 for flowing a liquid by machining a groove having a width of 5 mm and a depth of 5 mm on the surface of the aluminum plate 10. A groove having a width of 2 mm and a depth of 1 mm was formed outside the flow path 7 along the outer periphery of the aluminum plate 10, and an O-ring 8 was inserted. Further, through holes were formed at both ends of the flow channel 7 to provide a flow channel inlet 7a and a flow channel outlet 7b for supplying and discharging liquid. Furthermore, a plurality of through-holes 9 were formed in the aluminum plate 10 in order to insert an energizing wire for energizing the heater and thermocouples for measuring temperatures at a plurality of locations of the heater. The aluminum plate having a thickness of 7 mm was fixed to the aluminum plate 10 by screwing to form a cooling block having a flow path inside.

  The heater and the cooling block were installed in a container having a predetermined shape, and a drive shaft was attached to the cooling block so as to be movable. Further, an energization line and a plurality of thermocouples were attached to the heater through a through hole of the cooling block. Each example described below was performed using the semiconductor manufacturing apparatus configured as described above. The temperature distribution on the wafer mounting surface of the heater is measured by mounting a wafer thermometer on the wafer mounting surface, and the difference between the maximum value and the minimum value of the measured value of the wafer thermometer is defined as the temperature variation of the heater. . In addition, the time until the heater temperature reached from 400 ° C. to 50 ° C. was measured by a thermocouple during cooling, and used as an index of the cooling capacity of the cooling block.

[Example 1]
As the cooling block, as shown in Table 1 below, the cooling blocks of samples 1 to 10 each having a predetermined warp on the contact surface with which the heater contacts were used. At this time, when these cooling blocks come into contact with the heater, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 ° or less, and the corner portion of the contact surface of the cooling block is chamfered to 100 μm. Has been. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

  The heater was heated to 400 ° C. as measured by a thermocouple, and held at 400 ° C. for 30 minutes to stabilize the temperature. Then, the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. During this energization and cooling, no liquid is flowing through the cooling block. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reaches 250 ° C., the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C., and the cooling capacity from 400 ° C. to 50 ° C. Time was measured.

  The obtained results are shown in Table 1 below together with the warpage of the contact surface of the cooling block. It can be seen that the uniformity of the temperature and the cooling capacity on the wafer mounting surface of the heater are greatly improved when the warpage of the heater contact surface of the cooling block is 1.0 mm, 0.2 mm, and 0.05 mm.

[Example 2]
A cooling module in which the warpage of the contact surface with which the heater contacts is 0.03 mm and the corner portion of the contact surface is chamfered by 100 μm was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times. At this time, in Samples 11 to 14, the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block was in contact with the heater was set to the angle shown in Table 2 below.

  As in Example 1, the heater temperature was heated to 400 ° C. and held. The heater was cooled to 50 ° C. by stopping the energization and bringing the cooling block into contact. During this energization and cooling, no liquid is flowing through the cooling block. The temperature at the moment when the heater temperature reached 390 ° C. as measured by the thermocouple was measured, and the temperature variation on the wafer mounting surface at the moment when the cooling block contacted the heater was evaluated.

  The obtained results are shown in Table 2 below together with the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block comes into contact with the heater. It can be seen that when the angle formed between the heater back surface and the contact surface of the cooling block at the moment of contact between the heater and the cooling block is 10 ° or less, the temperature variation on the wafer mounting surface of the heater is greatly reduced.

[Example 3]
A cooling block having a chamfering process as shown in Table 3 below was used at the corner of the contact surface that contacted the heater of the cooling block. At this time, the warpage of the contact surface with which the heater of the cooling block contacts was 0.03 mm, and the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block was in contact with the heater was 5 ° or less. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

  As in Example 1, the heater temperature was heated to 400 ° C. and held. The heater was cooled to 50 ° C. by stopping the energization and bringing the cooling block into contact. During this energization and cooling, no liquid is flowing through the cooling block. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reaches 250 ° C., the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C., and the cooling capacity from 400 ° C. to 50 ° C. Time was measured.

  The obtained results are shown in Table 3 below together with the chamfering amount of the contact surface of the cooling block. It can be seen that when the chamfering amount is 10 μm or more, the uniformity of temperature and the cooling capacity on the wafer mounting surface of the heater are greatly improved.

[Example 4]
When the cooling block is separated from the heater and is stationary, a sample is used in which the angle formed between the contact surface of the cooling block and the back surface of the heater is an angle shown in Table 3 below. At this time, the warpage of the contact surface with which the heater of the cooling block contacts was 0.03 mm, and the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block was in contact with the heater was 5 ° or less. The chamfering process at the corner of the contact surface of the cooling block was 100 μm. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; In an intersection line intersecting with an arbitrary plane perpendicular to the placement surface and passing through the center of the wafer placement surface, the length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the length. The average value is adjusted to 0.97 to 1.03 times.

  After the heater was heated to 400 ° C. with a thermocouple measurement, the temperature was kept at 400 ° C. for 30 minutes, and the temperature variation of the wafer mounting surface was measured when the temperature was stabilized. Thereafter, the heater was cooled to 50 ° C. by stopping the energization and bringing the cooling block into contact therewith. During this energization and cooling, no liquid is flowing through the cooling block. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

  The obtained results are shown in Table 4 below together with the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block is separated from the heater and is stationary. When the angle between the abutment surface of the cooling block and the back surface of the heater is 10 ° or less when the cooling block is separated from the heater, the temperature variation at 400 ° C. is small, and even during cooling You can see that the effect remains.

[Example 5]
A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer mounting surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is an average of the lengths at an intersection line intersecting with an arbitrary plane passing through the center of the wafer mounting surface and perpendicular to the surface. Samples having the maximum variation shown in Table 5 below with respect to the values were prepared. At this time, the warpage of the contact surface with which the heater of the cooling block contacts is 0.03 mm, and the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block contacts the heater is 5 ° or less. The chamfering of the corner of the contact surface of the block was 100 μm, and the angle formed between the contact surface of the cooling block and the back surface of the heater when the cooling block was separated from the heater and was stationary was 5 °.

  After the heater was heated to 400 ° C. with a thermocouple measurement, the temperature was kept at 400 ° C. for 30 minutes, and the temperature variation of the wafer mounting surface was measured when the temperature was stabilized. Thereafter, the heater was cooled to 50 ° C. by stopping the energization and bringing the cooling block into contact therewith. During this energization and cooling, no liquid is flowing through the cooling block. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

  The obtained results are shown in Table 5 below. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer mounting surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is an average of the lengths at an intersection line intersecting with an arbitrary plane passing through the center of the wafer mounting surface and perpendicular to the surface. When it is in the range of 0.9 to 1.1 times the value, the temperature variation at 400 ° C. is small, and it can be seen that the effect remains even during cooling.

[Example 6]
The warpage of the contact surface with which the heater contacts is 0.02 mm, the angle formed between the contact surface of the cooling block and the back of the heater when contacting the heater is 5 °, and the corner portion of the contact surface is chamfered to 50 μm. A cooling block was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

In addition, a flow path is formed inside the cooling block, and the cross-sectional area of the flow path is 25 mm ² over the entire length of the flow path, and the flow area on the contact surface of the cooling block (perpendicular to the contact surface on which the heater contacts) When viewed from the direction, the same applies hereinafter) is 15% of the area of the contact surface. Further, the flow path is formed on the inner side of a position 20 mm from the outer edge of the contact surface of the cooling block (the same applies hereinafter when viewed from the direction perpendicular to the contact surface with which the heater contacts). The surface roughness of the surface that contacts is 0.2 μm in Ra.

  Water was passed through the cooling block at a flow rate shown in Table 6 below during heating, holding, and cooling of the heater. The heater temperature was heated to 400 ° C. at 20 ° C./minute and held at 400 ° C. for 30 minutes to stabilize the temperature, and then the heater was cooled to 50 ° C. by stopping the energization and contacting the cooling block. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C., the temperature variation at the moment of reaching 50 ° C., and the cooling capacity were measured. Further, at the moment when the temperature reached 50 ° C., the cooling block was separated from the heater and heated again to 400 ° C. at 20 ° C./min, and the cooling capacity was measured three times in total by repeating heating and cooling.

  The obtained results are shown in Table 6 below together with the amount of cooling water that has flowed through the flow path in the cooling block. When cooling water is flowed at a flow rate of 500 cc / min or more, the uniformity of the temperature and the cooling capacity on the wafer mounting surface of the heater are greatly improved, and the cooling capacity does not decrease even if the heating and cooling of the heater are repeated. I understand.

[Example 7]
The warpage of the contact surface that contacts the heater is 0.02 mm, the angle formed between the contact surface of the cooling block and the back of the heater when contacting the heater is 5 °, and the corner portion of the contact surface is chamfered to 50 μm. A cooling block was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

  In addition, a flow path is formed inside the cooling block, and the flow area on the contact surface of the cooling block is 15% of the area of the contact surface, and the flow path is at a position 20 mm from the outer edge of the contact surface. The surface roughness of the surface formed on the inside and in contact with the liquid in the flow path is 0.2 μm in Ra. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

As described above, the flow path in the cooling block has a width of 5 mm and a depth of 5 mm (cross-sectional area of 25 mm ² ). However, by changing the width and depth of a part of the flow path, the cross-sectional area with respect to the total length of the flow path is 1 mm ^2. A cooling block in which the ratio of the above flow paths is shown in Table 7 below was used. The heater was heated to 400 ° C. as measured by a thermocouple, held at 400 ° C. for 30 minutes to stabilize the temperature, and then the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

The obtained results are shown in Table 7 below together with the ratio of the flow path having a cross-sectional area of 1 mm ² or more with respect to the total length of the flow path. It can be seen that when the ratio of the flow path having a cross-sectional area of 1 mm ² or more with respect to the total length of the flow path is 80% or more, the temperature uniformity on the wafer mounting surface of the heater is greatly improved.

[Example 8]
The warpage of the contact surface that contacts the heater is 0.02 mm, the angle formed between the contact surface of the cooling block and the back of the heater when contacting the heater is 5 °, and the corner portion of the contact surface is chamfered to 50 μm. A cooling block was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

A flow path is formed inside the cooling block, the cross-sectional area of the flow path in the cooling block is 25 mm ² over the entire length of the flow path, and the flow path is formed on the inner side of the position 20 mm from the outer edge of the contact surface. The surface roughness of the surface in contact with the liquid in the channel is 0.2 μm in Ra. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

  When viewed from the direction perpendicular to the contact surface with which the heater of the cooling block contacts, the cooling block with the ratio of the flow path formed on the contact surface to the area of the contact surface shown in Table 8 below Was used. The heater was heated to 400 ° C. as measured by a thermocouple, held at 400 ° C. for 30 minutes to stabilize the temperature, and then the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

  The obtained results are shown in Table 8 below together with the ratio of the channel area to the contact surface area of the cooling block. It can be seen that the uniformity of temperature on the wafer mounting surface of the heater is greatly improved when the ratio of the flow path area to the contact surface area is 3% or more.

[Example 9]
The warpage of the contact surface that contacts the heater is 0.02 mm, the angle formed between the contact surface of the cooling block and the back of the heater when contacting the heater is 5 °, and the corner portion of the contact surface is chamfered to 50 μm. A cooling block was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

A flow path is formed inside the cooling block, the cross-sectional area of the flow path inside the cooling block is 25 mm ² over the entire length of the flow path, and the area where the flow path is formed is 15% of the area of the contact surface. The surface roughness of the surface in contact with the liquid in the channel is 0.2 μm in Ra. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

  A cooling block was used in which the shortest distance from the outer edge of the contact surface of the cooling block to the flow path was changed as shown in Table 9 below, when viewed from the direction perpendicular to the contact surface that contacts the heater of the cooling block. . The heater was heated to 400 ° C. as measured by a thermocouple, held at 400 ° C. for 30 minutes to stabilize the temperature, and then the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

  The obtained results are shown in Table 9 below together with the shortest distance from the outer edge of the contact surface of the cooling block to the flow path. It can be seen that when the shortest distance from the outer edge of the contact surface of the cooling block to the flow path is 50 mm or less, the temperature uniformity on the wafer mounting surface of the heater is greatly improved.

[Example 10]
The warpage of the contact surface that contacts the heater is 0.02 mm, the angle formed between the contact surface of the cooling block and the back of the heater when contacting the heater is 5 °, and the corner portion of the contact surface is chamfered to 50 μm. A cooling block was used. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

In addition, a flow path is formed inside the cooling block, the cross-sectional area of the flow path in the cooling block is 25 mm ² over the entire length of the flow path, and the area where the flow path is formed is the area of the contact surface 15%, and the flow path is formed on the inner side of the position 20 mm from the outer edge of the contact surface. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

  A cooling block in which the surface roughness (Ra) of the surface in contact with the liquid in the flow path of the cooling block was changed as shown in Table 10 below was used. The heater was heated to 400 ° C. as measured by a thermocouple, held at 400 ° C. for 30 minutes to stabilize the temperature, and then the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C., the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C., and the cooling capacity were measured.

  The obtained results are shown in Table 10 below together with the surface roughness (Ra) of the surface with which the liquid in the channel of the cooling block contacts. It can be seen that when the surface roughness of the surface in contact with the liquid in the flow path is in the range of 0.0 to 100 μm Ra, the uniformity of the temperature and the cooling capacity on the wafer mounting surface of the heater are greatly improved. .

[Example 11]
The aluminum nitride (AlN) heater is made of aluminum oxide (Al ₂ O ₃ ), silicon carbide (SiC), and silicon nitride (Si ₃ N ₄ ), and has the same shape as in the case of the aluminum nitride heater. Each heater was manufactured. On the other hand, in the cooling block, the warpage of the contact surface that contacts the heater is 0.02 mm, the angle formed between the contact surface of the cooling block and the heater back surface when contacting the heater is 5 °, and the corner of the contact surface is 50 μm. A cooling block with chamfering was used.

  When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

Further, a flow path is formed inside the cooling block, the cross-sectional area of the flow path in the cooling block is 25 mm ² over the entire length of the flow path, and the area where the flow path is formed is 15 of the area of the contact surface. The flow path is formed on the inner side of the position 20 mm from the outer edge of the contact surface, and the surface roughness of the surface in contact with the liquid in the flow path is 0.2 μm in Ra. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

  The heater was heated to 400 ° C. as measured by a thermocouple, held at 400 ° C. for 30 minutes to stabilize the temperature, and then the energization was stopped and the cooling block was contacted to cool the heater to 50 ° C. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured. In addition, the above repeated heating and cooling tests were repeated up to 1000 times, and the number of tests until the heater was damaged was examined.

  The obtained results are shown in Table 11 below together with the heater material. Aluminum nitride and silicon carbide are superior in uniformity of temperature on the wafer mounting surface of the heater, and aluminum nitride and silicon nitride are superior in reliability evaluated by the above repeated test. Among them, aluminum nitride is temperature uniformity and reliability. It turns out that it combines.

[Example 12]
In the same manner as the aluminum cooling block described above, cooling blocks made of five types of materials shown in Table 12 below were produced. In each cooling block, the warpage of the contact surface that contacts the heater is 0.02 mm, the angle between the contact surface of the cooling block and the heater back surface when contacting the heater is 5 °, and the corner of the contact surface is 50 μm. Chamfered. When the cooling block is separated from the heater and is stationary, the angle formed between the contact surface of the cooling block and the back surface of the heater is 5 °. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; and the wafer mounting The length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the average of the lengths at the intersection line intersecting with an arbitrary plane passing through the center of the wafer placement surface and perpendicular to the surface. The value is adjusted to 0.97 to 1.03 times.

In addition, a flow path is formed in each cooling block, the cross-sectional area of the flow path is 25 mm ² over the entire length of the flow path, and the area where the flow path is formed is 15% of the area of the contact surface. The channel has a channel formed inside 20 mm from the outer edge of the contact surface, and the surface roughness of the surface with which the liquid in the channel contacts is 0.2 μm in Ra. 1000 cc / min of water was allowed to flow through the flow path of the cooling block during heating, holding, and cooling of the heater.

  Heat the aluminum nitride heater to 400 ° C. as measured by a thermocouple, hold the temperature at 400 ° C. for 30 minutes to stabilize the temperature, and then bring the heater to 50 ° C. by stopping the energization and contacting the cooling block. Cooled down. The temperature variation of the wafer mounting surface at the moment when the measured value of the thermocouple reached 250 ° C. and the temperature variation of the wafer mounting surface at the moment of reaching 50 ° C. were measured.

  The obtained results are shown in Table 12 below together with the material of the cooling block and its thermal conductivity. It can be seen that the uniformity of temperature on the wafer mounting surface of the heater is improved when the thermal conductivity of the material constituting the cooling block is 30 W / mK and 100 W / mK.

It is a schematic sectional drawing of the semiconductor manufacturing apparatus of this invention, Comprising: The state which the heater and cooling block of the wafer holding body isolate | separated is shown. It is a schematic sectional drawing of the semiconductor manufacturing apparatus of this invention, Comprising: The state which the heater and cooling block of the wafer holder contact | abutted is shown. It is a schematic sectional drawing of the cooling block concerning this invention, Comprising: The state of the flow path formed in the inside is shown.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Heater 1a Wafer mounting surface 1b Heater back surface 2 Cooling block 2a Contact surface 3 Container 4 Thermocouple 5 Conducting wire 6 Drive shaft 7 Flow path 7a Flow path inlet 7b Flow path outlet 8 O-ring 9 Through hole 10 Aluminum plate

Claims (17)

A wafer holder comprising a heater for mounting and heating a semiconductor wafer and a cooling block for cooling the heater is provided, and the cooling block contacts and separates from the back surface of the heater opposite to the wafer mounting surface. A semiconductor manufacturing apparatus, wherein the warp of the contact surface which is arranged so as to be movable and which contacts the heater of the cooling block is 1 mm or less. The semiconductor manufacturing apparatus according to claim 1, wherein a warp of a contact surface that contacts the heater of the cooling block is 0.2 mm or less. 3. The semiconductor manufacturing apparatus according to claim 2, wherein a curvature of a contact surface that contacts the heater of the cooling block is 0.05 mm or less. The angle formed by the contact surface of the cooling block and the back surface of the heater when the cooling block moves and contacts the heater is 10 ° or less. Semiconductor manufacturing equipment. 5. The semiconductor manufacturing apparatus according to claim 1, wherein a chamfering process of 10 μm or more is applied to a corner portion of a contact surface that contacts the heater of the cooling block. The angle between the contact surface of the cooling block and the back surface of the heater is 10 ° or less when the cooling block is separated from the heater and is stationary. Semiconductor manufacturing equipment. A cylindrical surface having a center line perpendicular to the wafer mounting surface of the heater and passing through the center of the wafer mounting surface and having a radius shorter than the distance from the center of the wafer mounting surface to the outer edge; In an intersection line intersecting with an arbitrary plane perpendicular to the placement surface and passing through the center of the wafer placement surface, the length from the wafer placement surface on the intersection line to the inner surface of the container of the semiconductor manufacturing apparatus is the length of the length The semiconductor manufacturing apparatus according to claim 1, wherein the semiconductor manufacturing apparatus is in a range of 0.9 to 1.1 times the average value. The semiconductor manufacturing apparatus according to claim 1, further comprising a cooling liquid channel in the cooling block. 9. The semiconductor manufacturing apparatus according to claim 8, wherein 80% or more of the total length of the flow path is 1 mm ² or more in cross-sectional area of the flow path. The area of the portion where the flow path is formed is 3% or more of the total area of the contact surface when viewed from a direction perpendicular to the contact surface with which the heater of the cooling block contacts. The semiconductor manufacturing apparatus according to claim 8 or 9. The flow path is formed in a range inward of 50 mm from the outer edge of the contact surface when viewed from a direction perpendicular to the contact surface with which the heater of the cooling block contacts. The semiconductor manufacturing apparatus in any one of -10. 12. The semiconductor manufacturing apparatus according to claim 8, wherein the surface roughness of the surface of the flow path that contacts the liquid is 0.0 to 100 μm in terms of Ra. The semiconductor manufacturing apparatus according to claim 8, wherein a flow rate of the liquid flowing through the flow path is 500 cc / min or more. The semiconductor manufacturing apparatus according to any one of claims 1 to 13, wherein the material constituting the cooling block has a thermal conductivity of 30 W / mK or more. The semiconductor manufacturing apparatus according to claim 14, wherein the material constituting the cooling block has a thermal conductivity of 100 W / mK or more. The semiconductor manufacturing apparatus according to claim 1, wherein a main component of the material constituting the heater is any one of aluminum nitride, aluminum oxide, silicon carbide, and silicon nitride. The semiconductor manufacturing apparatus according to claim 16, wherein a material constituting the heater is mainly made of aluminum nitride.

Images